public FastVector GetDataSetAtts()
{
if (_fvWekaAttributes != null)
{
return(_fvWekaAttributes);
}
// Declare features
FastVector fvWekaAttributes = new FastVector(DomPool.SelectorFeatures.Count() + 1);
foreach (Feature currFeature in DomPool.SelectorFeatures)
{
weka.core.Attribute feature = new weka.core.Attribute(currFeature.ToString());
fvWekaAttributes.addElement(feature);
}
// Declare the class attribute along with its values
FastVector fvClassVal = new FastVector(2);
fvClassVal.addElement("yes");
fvClassVal.addElement("no");
weka.core.Attribute ClassAttribute = new weka.core.Attribute("theClass", fvClassVal);
// Declare the feature vector
fvWekaAttributes.addElement(ClassAttribute);
_fvWekaAttributes = fvWekaAttributes;
return(_fvWekaAttributes);
}
public Identifier(Boolean isAnnotated, WhyTrainer wt)
{
this.isAnnotated = isAnnotated;
this.wt = wt;
listWhoCandidates = new List <Candidate>();
listWhenCandidates = new List <Candidate>();
listWhereCandidates = new List <Candidate>();
listWhatCandidates = new List <List <Token> >();
listWhyCandidates = new List <List <Token> >();
listSecondaryWhyCandidates = new List <Candidate>();
fvPOS = new FastVector(Token.PartOfSpeechTags.Length);
foreach (String POS in Token.PartOfSpeechTags)
{
fvPOS.addElement(POS);
}
whoClassifier = (Classifier)SerializationHelper.read(@"..\..\IdentifierModels\who.model");
whenClassifier = (Classifier)SerializationHelper.read(@"..\..\IdentifierModels\when.model");
whereClassifier = (Classifier)SerializationHelper.read(@"..\..\IdentifierModels\where.model");
whyClassifier = (Classifier)SerializationHelper.read(@"..\..\IdentifierModels\why.model");
initializeAnnotations();
}
public HashSet <HtmlNode> RunOnTestSeenSet()
{
HashSet <HtmlNode> classifierSelectedNodes = new HashSet <HtmlNode>();
InitTestSeen();
foreach (string featureString in FeaturesUsed)
{
HashSet <HtmlNode> resNodes = DomPool.TESTSeenRunXpathQuery(useNormalPerformanceQUERY(featureString));
foreach (HtmlNode nd in resNodes)
{
if (!testSeenAllNodes.Contains(nd))
{
continue;
}
testSeenNodeFeatures[nd].Add(featureString);
}
}
FastVector fvWekaAttributes = GetDataSetAtts();
Instances testSet = new Instances("TestSeenSet", fvWekaAttributes, 10);
testSet.setClassIndex(fvWekaAttributes.size() - 1);
foreach (HtmlNode currNode in testSeenAllNodes)
{
Instance item = new SparseInstance(fvWekaAttributes.size());
for (int i = 0; i < fvWekaAttributes.size() - 1; i++)
{
weka.core.Attribute currFeature = (weka.core.Attribute)fvWekaAttributes.elementAt(i);
if (testSeenNodeFeatures[currNode].Contains(currFeature.name()))
{
item.setValue(currFeature, 1);
}
else
{
item.setValue(currFeature, 0);
}
}
//set the class
weka.core.Attribute classFeature = (weka.core.Attribute)fvWekaAttributes.elementAt(fvWekaAttributes.size() - 1);
//string rightVal = DomPool.TargetNodes.Contains(currNode) ? "yes" : "no";
item.setDataset(testSet);
double classifierdv = classifierTree.classifyInstance(item);
string classifierVal = classFeature.value((int)classifierdv);
if (classifierVal.Equals("yes"))
{
classifierSelectedNodes.Add(currNode);
}
testSet.add(item);
}
return(classifierSelectedNodes);
}
private FastVector createWhereFastVector()
{
FastVector fvWhere = new FastVector(5 + whereWordsBefore * 2 + whereWordsAfter * 2);
fvWhere.addElement(new weka.core.Attribute("word", (FastVector)null));
fvWhere.addElement(new weka.core.Attribute("wordCount"));
fvWhere.addElement(new weka.core.Attribute("sentence"));
//fvWhere.addElement(new weka.core.Attribute("position"));
//fvWhere.addElement(new weka.core.Attribute("sentenceStartProximity"));
fvWhere.addElement(new weka.core.Attribute("wordScore"));
for (int i = whereWordsBefore; i > 0; i--)
{
fvWhere.addElement(new weka.core.Attribute("word-" + i, (FastVector)null));
}
for (int i = 1; i <= whereWordsAfter; i++)
{
fvWhere.addElement(new weka.core.Attribute("word+" + i, (FastVector)null));
}
for (int i = whereWordsBefore; i > 0; i--)
{
fvWhere.addElement(new weka.core.Attribute("postag-" + i, fvPOS));
}
for (int i = 1; i <= whereWordsAfter; i++)
{
fvWhere.addElement(new weka.core.Attribute("postag+" + i, fvPOS));
}
FastVector fvClass = new FastVector(2);
fvClass.addElement("yes");
fvClass.addElement("no");
fvWhere.addElement(new weka.core.Attribute("where", fvClass));
return(fvWhere);
}
private static Instances CreateInstanceOnFly(double[] a, double[] b)
{
FastVector atts;
Instances data;
double[] vals;
// 1. set up attributes
atts = new FastVector();
// - numeric
atts.addElement(new Attribute("att1"));
atts.addElement(new Attribute("att2"));
// 2. create Instances object
data = new Instances("MyRelation", atts, 0);
for (int i = 0; i < a.Length; ++i)
{
// 3. fill with data
// first instance
vals = new double[data.numAttributes()];
// - numeric
vals[0] = a[i];
// - nominal
vals[1] = b[i];
data.add(new weka.core.DenseInstance(1.0, vals));
}
return(data);
}
private FastVector createWhyFastVector()
{
FastVector fvWhy = new FastVector(8 + whyWordsBefore * 2 + whyWordsAfter * 2);
fvWhy.addElement(new weka.core.Attribute("candidate", (FastVector)null));
fvWhy.addElement(new weka.core.Attribute("wordCount"));
fvWhy.addElement(new weka.core.Attribute("sentence"));
fvWhy.addElement(new weka.core.Attribute("candidateScore"));
fvWhy.addElement(new weka.core.Attribute("numWho"));
fvWhy.addElement(new weka.core.Attribute("numWhen"));
fvWhy.addElement(new weka.core.Attribute("numWhere"));
for (int i = whereWordsBefore; i > 0; i--)
{
fvWhy.addElement(new weka.core.Attribute("word-" + i, (FastVector)null));
}
for (int i = 1; i <= whereWordsAfter; i++)
{
fvWhy.addElement(new weka.core.Attribute("word+" + i, (FastVector)null));
}
for (int i = whyWordsBefore; i > 0; i--)
{
fvWhy.addElement(new weka.core.Attribute("postag-" + i, fvPOS));
}
for (int i = 1; i <= whyWordsAfter; i++)
{
fvWhy.addElement(new weka.core.Attribute("postag+" + i, fvPOS));
}
FastVector fvClass = new FastVector(2);
fvClass.addElement("yes");
fvClass.addElement("no");
fvWhy.addElement(new weka.core.Attribute("why", fvClass));
return(fvWhy);
}
private Instance createSingleWhoInstance(FastVector fvWho, Token candidate)
{
//first word-n attribute number
int wordsBeforeFirstAttributeNumber = 6;
//first pos-n attribute number
int posBeforeFirstAttributeNumber = wordsBeforeFirstAttributeNumber + whoWordsBefore + whoWordsAfter;
//word+1 attribute number
int wordsAfterFirstAttributeNumber = wordsBeforeFirstAttributeNumber + whoWordsBefore;
//pos+1 attribute number
int posAfterFirstAttributeNumber = posBeforeFirstAttributeNumber + whoWordsBefore;
int totalAttributeCount = wordsBeforeFirstAttributeNumber + whoWordsBefore * 2 + whoWordsAfter * 2 + 1;
Instance whoCandidate = new DenseInstance(totalAttributeCount);
whoCandidate.setValue((weka.core.Attribute)fvWho.elementAt(0), candidate.Value);
whoCandidate.setValue((weka.core.Attribute)fvWho.elementAt(1), candidate.Value.Split(' ').Count());
whoCandidate.setValue((weka.core.Attribute)fvWho.elementAt(2), candidate.Sentence);
whoCandidate.setValue((weka.core.Attribute)fvWho.elementAt(3), candidate.Position);
double sentenceStartProximity = -1;
foreach (List <Token> tokenList in segregatedArticleCurrent)
{
if (tokenList.Count > 0 && tokenList[0].Sentence == candidate.Sentence)
{
sentenceStartProximity = (double)(candidate.Position - tokenList[0].Position) / (double)tokenList.Count;
break;
}
}
if (sentenceStartProximity > -1)
{
whoCandidate.setValue((weka.core.Attribute)fvWho.elementAt(4), sentenceStartProximity);
}
whoCandidate.setValue((weka.core.Attribute)fvWho.elementAt(5), candidate.Frequency);
for (int i = whoWordsBefore; i > 0; i--)
{
if (candidate.Position - i - 1 >= 0)
{
whoCandidate.setValue((weka.core.Attribute)fvWho.elementAt(whoWordsBefore - i + wordsBeforeFirstAttributeNumber), articleCurrent[candidate.Position - i - 1].Value);
if (articleCurrent[candidate.Position - i - 1].PartOfSpeech != null)
{
whoCandidate.setValue((weka.core.Attribute)fvWho.elementAt(whoWordsBefore - i + posBeforeFirstAttributeNumber), articleCurrent[candidate.Position - i - 1].PartOfSpeech);
}
}
}
for (int i = 0; i < whoWordsAfter; i++)
{
if (candidate.Position + i < articleCurrent.Count)
{
whoCandidate.setValue((weka.core.Attribute)fvWho.elementAt(wordsAfterFirstAttributeNumber + i), articleCurrent[candidate.Position + i].Value);
if (articleCurrent[candidate.Position + i].PartOfSpeech != null)
{
whoCandidate.setValue((weka.core.Attribute)fvWho.elementAt(posAfterFirstAttributeNumber + i), articleCurrent[candidate.Position + i].PartOfSpeech);
}
}
}
return(whoCandidate);
}
private static FastVector PrepareFastVector()
{
var fastVector = new FastVector(_attributes.Count);
foreach (var attribute in _attributes)
{
fastVector.addElement(attribute);
}
return(fastVector);
}
private static Instances DefineBucketAttributes(int numBuckets)
{
FastVector attributes = new FastVector();
FastVector numberOfTimesPregnant = new FastVector();
FastVector plasmaGlucoseConcentration = new FastVector();
FastVector diastolicBloodPressure = new FastVector();
FastVector tricepsSkinFoldThickness = new FastVector();
FastVector twoHourSerumInsulin = new FastVector();
FastVector bmi = new FastVector();
FastVector diabetesPedigreeFunction = new FastVector();
FastVector age = new FastVector();
for (int i = 0; i < numBuckets; i++)
{
string attributeValueName = i.ToString();
numberOfTimesPregnant.addElement(attributeValueName);
plasmaGlucoseConcentration.addElement(attributeValueName);
diastolicBloodPressure.addElement(attributeValueName);
tricepsSkinFoldThickness.addElement(attributeValueName);
twoHourSerumInsulin.addElement(attributeValueName);
bmi.addElement(attributeValueName);
diabetesPedigreeFunction.addElement(attributeValueName);
age.addElement(attributeValueName);
}
attributes.addElement(new weka.core.Attribute("numberOfTimesPregnant", numberOfTimesPregnant));
attributes.addElement(new weka.core.Attribute("plasmaGlucoseConcentration", plasmaGlucoseConcentration));
attributes.addElement(new weka.core.Attribute("diastolicBloodPressure", diastolicBloodPressure));
attributes.addElement(new weka.core.Attribute("tricepsSkinFoldThickness", tricepsSkinFoldThickness));
attributes.addElement(new weka.core.Attribute("twoHourSerumInsulin", twoHourSerumInsulin));
attributes.addElement(new weka.core.Attribute("bmi", bmi));
attributes.addElement(new weka.core.Attribute("diabetesPedigreeFunction", diabetesPedigreeFunction));
attributes.addElement(new weka.core.Attribute("age", age));
FastVector diabetes = new FastVector();
diabetes.addElement("0"); // negative
diabetes.addElement("1"); // positive
attributes.addElement(new weka.core.Attribute("diagnosis", diabetes));
Instances instances = new Instances("diabetes", attributes, 0);
return(instances);
}
private Instance createSingleWhyInstance(FastVector fvWhy, Token candidate)
{
//first word-n attribute number
int wordsBeforeFirstAttributeNumber = 7;
//first pos-n attribute number
int posBeforeFirstAttributeNumber = wordsBeforeFirstAttributeNumber + whyWordsBefore + whyWordsAfter;
//word+1 attribute number
int wordsAfterFirstAttributeNumber = wordsBeforeFirstAttributeNumber + whyWordsBefore;
//pos+1 attribute number
int posAfterFirstAttributeNumber = posBeforeFirstAttributeNumber + whyWordsBefore;
int totalAttributeCount = wordsBeforeFirstAttributeNumber + whyWordsBefore * 2 + whyWordsAfter * 2 + 1;
Instance whyCandidate = new DenseInstance(totalAttributeCount);
whyCandidate.setValue((weka.core.Attribute)fvWhy.elementAt(0), candidate.Value);
whyCandidate.setValue((weka.core.Attribute)fvWhy.elementAt(1), candidate.Value.Split(' ').Count());
whyCandidate.setValue((weka.core.Attribute)fvWhy.elementAt(2), candidate.Sentence);
whyCandidate.setValue((weka.core.Attribute)fvWhy.elementAt(3), candidate.Score);
whyCandidate.setValue((weka.core.Attribute)fvWhy.elementAt(4), candidate.NumWho);
whyCandidate.setValue((weka.core.Attribute)fvWhy.elementAt(5), candidate.NumWhen);
whyCandidate.setValue((weka.core.Attribute)fvWhy.elementAt(6), candidate.NumWhere);
for (int i = whyWordsBefore; i > 0; i--)
{
if (candidate.Position - i - 1 >= 0)
{
whyCandidate.setValue((weka.core.Attribute)fvWhy.elementAt(whyWordsBefore - i + wordsBeforeFirstAttributeNumber), articleCurrent[candidate.Position - i - 1].Value);
if (articleCurrent[candidate.Position - i - 1].PartOfSpeech != null)
{
whyCandidate.setValue((weka.core.Attribute)fvWhy.elementAt(whyWordsBefore - i + posBeforeFirstAttributeNumber), articleCurrent[candidate.Position - i - 1].PartOfSpeech);
}
}
}
for (int i = 0; i < whyWordsAfter; i++)
{
if (candidate.Position + i < articleCurrent.Count)
{
whyCandidate.setValue((weka.core.Attribute)fvWhy.elementAt(wordsAfterFirstAttributeNumber + i), articleCurrent[candidate.Position + i].Value);
if (articleCurrent[candidate.Position + i].PartOfSpeech != null)
{
whyCandidate.setValue((weka.core.Attribute)fvWhy.elementAt(posAfterFirstAttributeNumber + i), articleCurrent[candidate.Position + i].PartOfSpeech);
}
}
}
return(whyCandidate);
}
public static List <ClusteredDataObject> Clustering(RoughSetInformations roughSetInformations, List <DataObject> dataObjects)
{
_dataObjects = dataObjects;
_attributes = roughSetInformations.ArgumentNames.Select(argumentName => new Attribute(argumentName)).ToList();
_fastVector = PrepareFastVector();
if (_dataObjects == null)
{
return(null);
}
_argumentsClustersRangeList = PrepareArgumentsClustersRangeList();
SortArgumentsClustersRangeListItems();
return(PrepareClusteredDataObjectsCollection());
}
int distanceComparer(Light a, Light b)
{
Vector3 posA = a.transform.position;
Vector3 posB = b.transform.position;
float distA = FastVector.SqrDistance(ref camPos, ref posA);
float distB = FastVector.SqrDistance(ref camPos, ref posB);
if (distA < distB)
{
return(-1);
}
if (distA > distB)
{
return(1);
}
return(0);
}
private Instances createWhyInstances()
{
FastVector fvWhy = createWhyFastVector();
Instances whyInstances = new Instances("WhyInstances", fvWhy, listSecondaryWhyCandidates.Count);
foreach (Token candidate in listSecondaryWhyCandidates)
{
if (candidate.Value == null)
{
continue;
}
Instance whyInstance = createSingleWhyInstance(fvWhy, candidate);
whyInstance.setDataset(whyInstances);
whyInstances.add(whyInstance);
}
whyInstances.setClassIndex(fvWhy.size() - 1);
return(whyInstances);
}
/// <summary>
/// Creates and returns empty weka data set </summary>
/// <param name="numOfAttr"> int Number of attributes without class attribute </param>
/// <param name="capacity"> int Capacity of sample </param>
/// <returns> empty weka data set </returns>
private static Instances createEmptyWekaDataSet(int numOfAttr, int capacity, IDictionary <double[], string> classValues)
{
//Vector for class attribute possible values
FastVector fvClassVal = new FastVector();
//Map double value for every possible class value
Hashtable classVals = new Dictionary <string, double?>();
//Map class label with double key value
Hashtable classValsDoubleAsKey = new Dictionary <double?, string>();
//ind represents double value for class attribute
int ind = 0;
//loop through possible class values
foreach (KeyValuePair <double[], string> values in classValues)
{
//add value to vector
fvClassVal.addElement(values.Value);
//map double value for class value
classVals[values.Value] = new double?(ind);
//map class label for double key value
classValsDoubleAsKey[new double?(ind)] = values.Value;
ind++;
}
//Class attribute with possible values
Attribute classAttribute = new Attribute("theClass", fvClassVal, classValues.Count);
//Creating attribute vector for Instances class instance
FastVector fvWekaAttributes = new FastVector(numOfAttr + 1);
//Fill vector with simple attributes
for (int i = 0; i < numOfAttr; i++)
{
fvWekaAttributes.addElement(new Attribute(i + "", i));
}
//Add class attribute to vector
fvWekaAttributes.addElement(classAttribute);
//newDataSet as Instances class instance
Instances newDataSet = new Instances("newDataSet", fvWekaAttributes, capacity);
return(newDataSet);
}
///
/// <summary> * Sets the format of output instances. </summary>
///
private void setOutputFormat()
{
// give names to the new attributes
FastVector newAtts = new FastVector();
string foName = null;
for (int i = 0; i < getInputFormat().numAttributes(); i++)
{
string attName = getInputFormat().attribute(i).name();
foName = "'FT " + attName.Replace('\'', ' ').Trim() + " (re)\'";
Attribute newAttribX = new Attribute(foName);
newAtts.addElement(newAttribX);
foName = "'FT " + attName.Replace('\'', ' ').Trim() + " (im)\'";
Attribute newAttribY = new Attribute(foName);
newAtts.addElement(newAttribY);
}
setOutputFormat(new Instances(getInputFormat().relationName(), newAtts, getNumCoeffs()));
}
//code taken from here http://stackoverflow.com/questions/9616872/classification-of-instances-in-weka/14876081#14876081
// This creates the data set's attributes vector
public static FastVector CreateFastVector(int size)
{
var fv = new FastVector();
weka.core.Attribute att;
foreach (int key in TrainingTesting_SharedVariables._trainTopIGFeatures)
{
if (key != TrainingTesting_SharedVariables._trainTopIGFeatures[TrainingTesting_SharedVariables._trainTopIGFeatures.Length - 1])
{
att = new weka.core.Attribute("att_" + (key + 1).ToString());
fv.addElement(att);
}
}
{
var classValues = new FastVector(1); //it doesnt matter if its 3 or 1, when addElement is used the fastvector grows.
List <string> labels = GuiPreferences.Instance.getLabels();
GuiPreferences.Instance.setLog("automatically! adding " + (labels.Count - 1).ToString() + " classes 2 -> " + (labels.Count + 1).ToString() + " to fast vector, based on protocol labels");
//baseline is ignored, we start from the second event in the protocol
for (int l = 1; l < labels.Count; l++)
{
classValues.addElement((l + 1).ToString());
}
//classValues.addElement("2");
//classValues.addElement("3");
//classValues.addElement("4");
//classValues.addElement("5");
var classAttribute = new weka.core.Attribute("class", classValues);
fv.addElement(classAttribute);
}
return(fv);
}
void RebuildCellRenderingLists(BatchedCell cell, Vector3 observerPos, float visibleDistance)
{
// rebuild batch lists to be used in the rendering loop
cell.ClearBatches();
float cullDistance = (visibleDistance * VoxelPlayEnvironment.CHUNK_SIZE) * (visibleDistance * VoxelPlayEnvironment.CHUNK_SIZE);
for (int j = 0; j < cell.instancedChunks.count; j++)
{
InstancedChunk instancedChunk = cell.instancedChunks.values [j];
if (instancedChunk == null)
{
continue;
}
// check if chunk is in area
Vector3 chunkCenter = instancedChunk.chunk.position;
if (FastVector.SqrDistance(ref chunkCenter, ref observerPos) > cullDistance)
{
continue;
}
// add instances to batch
InstancedVoxel[] voxels = instancedChunk.instancedVoxels.values;
for (int i = 0; i < instancedChunk.instancedVoxels.count; i++)
{
BatchedMesh batchedMesh = voxels [i].batchedMesh;
Batch batch = batchedMesh.batches.last;
if (batch == null || batch.instancesCount >= Batch.MAX_INSTANCES)
{
batch = batchedMesh.batches.FetchDirty();
if (batch == null)
{
batch = new Batch();
batch.instancedMaterial = GameObject.Instantiate <Material> (batchedMesh.material);
if (batchedMesh.voxelDefinition.rotationRandomY || batchedMesh.voxelDefinition.rotation != Misc.vector3zero)
{
batch.instancedMaterial.EnableKeyword(SKW_VOXELPLAY_USE_ROTATION);
}
batchedMesh.batches.Add(batch);
}
batch.Init();
}
int pos = batch.instancesCount++;
batch.positions [pos] = voxels [i].position;
batch.rotations [pos].x = voxels [i].rotation.x; batch.rotations [pos].y = voxels [i].rotation.y; batch.rotations [pos].z = voxels [i].rotation.z; batch.rotations [pos].w = voxels [i].rotation.w;
// batch.scales[pos] = voxels[i].scale;
batch.colorsAndLight [pos].x = voxels [i].color.r / 255f;
batch.colorsAndLight [pos].y = voxels [i].color.g / 255f;
batch.colorsAndLight [pos].z = voxels [i].color.b / 255f;
batch.colorsAndLight [pos].w = voxels [i].packedLight;
batch.UpdateBounds(voxels [i].position, voxels [i].meshSize);
}
}
for (int i = 0; i <= cell.batchedMeshes.lastIndex; i++)
{
BatchedMesh batchedMesh = cell.batchedMeshes.values [i];
if (batchedMesh == null)
{
continue;
}
for (int j = 0; j < batchedMesh.batches.count; j++)
{
Batch batch = batchedMesh.batches.values [j];
batch.ComputeBounds();
// Set positions
batch.positionsBuffer.SetData(batch.positions);
batch.instancedMaterial.SetBuffer("_Positions", batch.positionsBuffer);
// Set colors and light
batch.colorsAndLightBuffer.SetData(batch.colorsAndLight);
batch.instancedMaterial.SetBuffer("_ColorsAndLight", batch.colorsAndLightBuffer);
// Set rotations
batch.rotationsBuffer.SetData(batch.rotations);
batch.instancedMaterial.SetBuffer("_Rotations", batch.rotationsBuffer);
// Set buffer args
Mesh mesh = batchedMesh.voxelDefinition.mesh;
batch.args [0] = mesh.GetIndexCount(0);
batch.args [1] = (uint)batch.instancesCount;
batch.args [2] = mesh.GetIndexStart(0);
batch.args [3] = 0; // (uint)mesh.GetBaseVertex (0);
batch.argsBuffer.SetData(batch.args);
}
}
}
public void LearnModel()
{
Init();
foreach (Feature currFeature in DomPool.SelectorFeatures)
{
String featureString = currFeature.ToString();
HashSet <HtmlNode> resNodes = DomPool.RunXpathQuery(featureString);
foreach (HtmlNode nd in resNodes)
{
if (!allNodes.Contains(nd))
{
continue;
}
nodeFeatures[nd].Add(featureString);
}
}
FastVector fvWekaAttributes = GetDataSetAtts();
Instances trainingSet = new Instances("TS", fvWekaAttributes, 100);
trainingSet.setClassIndex(fvWekaAttributes.size() - 1);
foreach (HtmlNode currNode in allNodes)
{
Instance item = new SparseInstance(fvWekaAttributes.size());
for (int i = 0; i < fvWekaAttributes.size() - 1; i++)
{
weka.core.Attribute currFeature = (weka.core.Attribute)fvWekaAttributes.elementAt(i);
if (nodeFeatures[currNode].Contains(currFeature.name()))
{
item.setValue(currFeature, 1);
}
else
{
item.setValue(currFeature, 0);
}
}
//set the class
weka.core.Attribute classFeature = (weka.core.Attribute)fvWekaAttributes.elementAt(fvWekaAttributes.size() - 1);
item.setValue(classFeature, (DomPool.TargetNodes.Contains(currNode)?"yes":"no"));
item.setDataset(trainingSet);
if (DomPool.TargetNodes.Contains(currNode))
{
for (int t = 0; t < (DomPool.NonTargetNodes.Count() / DomPool.TargetNodes.Count()); t++)
{
trainingSet.add(new SparseInstance(item));
}
}
else
{
trainingSet.add(item);
}
}
String[] options = new String[2];
options[0] = "-C"; // unpruned tree
options[1] = "0.1";
J48 tree = new J48(); // new instance of tree
tree.setOptions(options); // set the options
tree.buildClassifier(trainingSet); // build classifier
//save the resulting classifier
classifierTree = tree;
Reader treeDot = new StringReader(tree.graph());
TreeBuild treeBuild = new TreeBuild();
Node treeRoot = treeBuild.create(treeDot);
FeaturesUsed = getTreeFeatures(treeRoot);
}
void RebuildZoneRenderingLists(Vector3 observerPos, float visibleDistance)
{
// rebuild batch lists to be used in the rendering loop
for (int k = 0; k < batchedMeshes.count; k++)
{
BatchedMesh batchedMesh = batchedMeshes.values [k];
batchedMesh.batches.Clear();
}
float cullDistance = (visibleDistance * VoxelPlayEnvironment.CHUNK_SIZE) * (visibleDistance * VoxelPlayEnvironment.CHUNK_SIZE);
for (int j = 0; j <= instancedChunks.lastIndex; j++)
{
InstancedChunk instancedChunk = instancedChunks.values [j];
if (instancedChunk == null)
{
continue;
}
// check if chunk is in area
Vector3 chunkCenter = instancedChunk.chunk.position;
if (FastVector.SqrDistance(ref chunkCenter, ref observerPos) > cullDistance)
{
continue;
}
// add instances to batch
InstancedVoxel[] voxels = instancedChunk.instancedVoxels.values;
for (int i = 0; i < instancedChunk.instancedVoxels.count; i++)
{
VoxelDefinition vd = voxels [i].voxelDefinition;
BatchedMesh batchedMesh = batchedMeshes.values [vd.batchedIndex];
Batch batch = batchedMesh.batches.last;
if (batch == null || batch.instancesCount >= Batch.MAX_INSTANCES)
{
batch = batchedMesh.batches.FetchDirty();
if (batch == null)
{
batch = new Batch();
batchedMesh.batches.Add(batch);
}
batch.Init();
}
int pos = batch.instancesCount++;
// just copying the matrix triggers lot of expensive memcpy() calls so we directly copy the fields
// batch.matrices[pos] = voxels[i].matrix;
batch.matrices [pos].m00 = voxels [i].matrix.m00; batch.matrices [pos].m01 = voxels [i].matrix.m01; batch.matrices [pos].m02 = voxels [i].matrix.m02; batch.matrices [pos].m03 = voxels [i].matrix.m03;
batch.matrices [pos].m10 = voxels [i].matrix.m10; batch.matrices [pos].m11 = voxels [i].matrix.m11; batch.matrices [pos].m12 = voxels [i].matrix.m12; batch.matrices [pos].m13 = voxels [i].matrix.m13;
batch.matrices [pos].m20 = voxels [i].matrix.m20; batch.matrices [pos].m21 = voxels [i].matrix.m21; batch.matrices [pos].m22 = voxels [i].matrix.m22; batch.matrices [pos].m23 = voxels [i].matrix.m23;
batch.matrices [pos].m30 = voxels [i].matrix.m30; batch.matrices [pos].m31 = voxels [i].matrix.m31; batch.matrices [pos].m32 = voxels [i].matrix.m32; batch.matrices [pos].m33 = voxels [i].matrix.m33;
batch.colorsAndLight [pos].x = voxels [i].color.r / 255f;
batch.colorsAndLight [pos].y = voxels [i].color.g / 255f;
batch.colorsAndLight [pos].z = voxels [i].color.b / 255f;
batch.colorsAndLight [pos].w = voxels [i].packedLight;
batch.UpdateBounds(voxels[i].position, voxels[i].meshSize);
}
}
for (int k = 0; k < batchedMeshes.count; k++)
{
BatchedMesh batchedMesh = batchedMeshes.values [k];
for (int j = 0; j < batchedMesh.batches.count; j++)
{
Batch batch = batchedMesh.batches.values [j];
batch.materialPropertyBlock.SetVectorArray("_TintColor", batch.colorsAndLight);
}
}
}
public void LearnModel()
{
Init();
foreach (Feature currFeature in DomPool.SelectorFeatures)
{
String featureString = currFeature.ToString();
HashSet <HtmlNode> resNodes = DomPool.RunXpathQuery(featureString);
foreach (HtmlNode nd in resNodes)
{
if (!allNodes.Contains(nd))
{
continue;
}
nodeFeatures[nd].Add(featureString);
}
}
FastVector fvWekaAttributes = GetDataSetAtts();
Instances trainingSet = new Instances("TS", fvWekaAttributes, 10);
trainingSet.setClassIndex(fvWekaAttributes.size() - 1);
foreach (HtmlNode currNode in allNodes)
{
Instance item = new SparseInstance(fvWekaAttributes.size());
for (int i = 0; i < fvWekaAttributes.size() - 1; i++)
{
weka.core.Attribute currFeature = (weka.core.Attribute)fvWekaAttributes.elementAt(i);
if (nodeFeatures[currNode].Contains(currFeature.name()))
{
item.setValue(currFeature, 1);
}
else
{
item.setValue(currFeature, 0);
}
}
//set the class
weka.core.Attribute classFeature = (weka.core.Attribute)fvWekaAttributes.elementAt(fvWekaAttributes.size() - 1);
item.setValue(classFeature, (DomPool.TargetNodes.Contains(currNode)?"yes":"no"));
item.setDataset(trainingSet);
if (DomPool.TargetNodes.Contains(currNode))
{
for (int t = 0; t < (DomPool.NonTargetNodes.Count() / DomPool.TargetNodes.Count()); t++)
{
trainingSet.add(new SparseInstance(item));
}
}
else
{
trainingSet.add(item);
}
}
//String[] options = new String[2];
//options = new string[] { "-C", "0.05" }; // unpruned tree
NaiveBayes cls = new NaiveBayes(); // new instance of tree
//cls.setOptions(weka.core.Utils.splitOptions("-C 1.0 -L 0.0010 -P 1.0E-12 -N 0 -V -1 -W 1 -K \"weka.classifiers.functions.supportVector.PolyKernel -C 250007 -E 1.0\""));
//cls.setOptions(options); // set the options
cls.buildClassifier(trainingSet); // build classifier
//save the resulting classifier
classifier = cls;
// Reader treeDot = new StringReader(tree.graph());
// TreeBuild treeBuild = new TreeBuild();
// Node treeRoot = treeBuild.create(treeDot);
FeaturesUsed = new HashSet <string>();
foreach (Feature f in DomPool.SelectorFeatures)
{
FeaturesUsed.Add(f.ToString());
}
}
// Define all the attributes for the diabetes dataset
private static Instances DefineCategoricalAttributes()
{
FastVector attributes = new FastVector();
FastVector numberOfTimesPregnant = new FastVector();
numberOfTimesPregnant.addElement("zero"); // 0
numberOfTimesPregnant.addElement("low"); // 1-4
numberOfTimesPregnant.addElement("medium"); // 5-9
numberOfTimesPregnant.addElement("high"); // 10+
attributes.addElement(new weka.core.Attribute("numberOfTimesPregnant", numberOfTimesPregnant));
FastVector plasmaGlucoseConcentration = new FastVector();
plasmaGlucoseConcentration.addElement("normal"); // < 140
plasmaGlucoseConcentration.addElement("high"); // >= 140
attributes.addElement(new weka.core.Attribute("plasmaGlucoseConcentration", plasmaGlucoseConcentration));
FastVector diastolicBloodPressure = new FastVector();
diastolicBloodPressure.addElement("low"); // < 60
diastolicBloodPressure.addElement("normal"); // 60 to 79
diastolicBloodPressure.addElement("pre-high"); // 80 to 89
diastolicBloodPressure.addElement("high"); // 90+
attributes.addElement(new weka.core.Attribute("diastolicBloodPressure", diastolicBloodPressure));
FastVector tricepsSkinFoldThickness = new FastVector();
tricepsSkinFoldThickness.addElement("low"); // < 4.5
tricepsSkinFoldThickness.addElement("normal"); // 4.5 to 36.5
tricepsSkinFoldThickness.addElement("high"); // > 36.5
attributes.addElement(new weka.core.Attribute("tricepsSkinFoldThickness", tricepsSkinFoldThickness));
FastVector twoHourSerumInsulin = new FastVector();
twoHourSerumInsulin.addElement("normal"); // < 166
twoHourSerumInsulin.addElement("high"); // >= 166
attributes.addElement(new weka.core.Attribute("twoHourSerumInsulin", twoHourSerumInsulin));
FastVector bmi = new FastVector();
bmi.addElement("underweight"); // < 18.5
bmi.addElement("normal"); // 18.5 to 25
bmi.addElement("overweight"); // 25 to 30
bmi.addElement("obese"); // 30+
attributes.addElement(new weka.core.Attribute("bmi", bmi));
FastVector diabetesPedigreeFunction = new FastVector();
diabetesPedigreeFunction.addElement("low"); // < 0.2
diabetesPedigreeFunction.addElement("normal"); // 0.2 to 0.7
diabetesPedigreeFunction.addElement("high"); // > 0.7
attributes.addElement(new weka.core.Attribute("diabetesPedigreeFunction", diabetesPedigreeFunction));
FastVector age = new FastVector();
age.addElement("young"); // under 30
age.addElement("middle"); // 30 to 50 inclusive
age.addElement("old"); // over 50
attributes.addElement(new weka.core.Attribute("age", age));
FastVector diabetes = new FastVector();
diabetes.addElement("0"); // negative
diabetes.addElement("1"); // positive
attributes.addElement(new weka.core.Attribute("diagnosis", diabetes));
Instances instances = new Instances("diabetes", attributes, 0);
return(instances);
}